Messages in this thread | | | From | Andi Kleen <> | Subject | Re: [Patch 2/3] fast VMA recycling | Date | Thu, 23 Feb 2006 12:00:42 +0100 |
| |
On Thursday 23 February 2006 11:15, Arjan van de Ven wrote: > On Thu, 2006-02-23 at 11:05 +0100, Andi Kleen wrote: > > On Thursday 23 February 2006 10:48, Arjan van de Ven wrote: > > > On Thu, 2006-02-23 at 10:42 +0100, Andi Kleen wrote: > > > > On Thursday 23 February 2006 10:30, Arjan van de Ven wrote: > > > > > This patch adds a per task-struct cache of a free vma. > > > > > > > > > > In normal operation, it is a really common action during userspace mmap > > > > > or malloc to first allocate a vma, and then find out that it can be merged, > > > > > and thus free it again. In fact this is the case roughly 95% of the time. > > > > > > > > > > In addition, this patch allows code to "prepopulate" the cache, and > > > > > this is done as example for the x86_64 mmap codepath. The advantage of this > > > > > prepopulation is that the memory allocation (which is a sleeping operation > > > > > due to the GFP_KERNEL flag, potentially causing either a direct sleep or a > > > > > voluntary preempt sleep) will happen before the mmap_sem is taken, and thus > > > > > reduces lock hold time (and thus the contention potential) > > > > > > > > The slab fast path doesn't sleep. > > > > > > it does via might_sleep() > > > > Hmm? That shouldn't sleep. > > see voluntary preempt.
Only when its time slice is used up but then it would sleep a bit later in user space. But it should be really a unlikely case and nothing to optimize for.
> > > If it takes any time in a real workload then it should move into DEBUG_KERNEL > > too. But I doubt it. Something with your analysis is wrong. > > well I'm seeing contention; and this is one of the things that can be > moved out of the lock easily, and especially given the high recycle rate > of these things... looks worth it to me.
I think you need a better analysis of what is actually happening instead of trying all kind of weird quick fragile hacks.
-Andi
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |